How to Develop Bulletproof Solutions to Complex Problems. Tips from Management Consulting.

Key takeaways

  • Define the problem. Frame up a problem statement that is specific, measurable, time bound and considers the values of the decision maker.
  • Break it down. Use an “issue tree” to visually map the main components of the problem.
  • Prioritise the issues. Target your work on the high impact, highly impactable parts of the tree.
  • Plan the work. Do the work. Develop a highly specific plan of who does what.
  • Synthesise findings and tell a great story. Use the “pyramid” principle to sell your solution.

How to Develop Bulletproof Solutions to Complex Problems. Tips from Management Consulting.

Most school improvement initiatives come down to problem-solving, you’re trying to improve something that isn’t working as well as it should. It’s not always easy to do this, and we know that leading school change isn’t just as simple as coming up with an idea and telling people to do it. Even before you get to that implementation stuff, you need to know that the solution you and your team have developed is a good one. Often trial and error, guesswork and assumptions play too big a role in developing solutions to complex problems, and this undermines how effective school leaders are in their improvement efforts.

Management consultants have made a business of being good at problem-solving. Over the past century companies like Mckinsey, BCG and Bain have built a multibillion-dollar industry around solving complex problems for business and governments. The biggest piece of value these companies offer their clients is a world-class problem-solving approach, and this is a learnable set of skills.

In their recent book Bulletproof Problem Solving: The One Skill That Changes Everything, two former Mckinsey partners Rob Mclean and Charles Conn provide the systematic 7-step problem-solving process used by leading consulting firms.

The Bulletproof framework is most useful when you need to make decisions that involve complexity and uncertainty (so there aren’t easy answers), and where there are consequences that make it worth investing time in getting good answers – sounds like a lot of problems in schools, right?

The framework is a highly visual and logical approach to problem-solving that can be applied anywhere. At its core, the framework is about defining problems, disaggregating them into manageable pieces, using good analytic tools on the most important parts of the problem and then synthesising your findings to tell a useful story.  

Let’s dig into this to work out exactly what you need to do to be a bulletproof problem solver.

Define the problem

It’s easy to waste a lot of time and energy solving the wrong problem or solving it in the wrong way, so correctly defining your problem is a critical element of success. Good problem definition can properly frame up what you’re trying to fix, and the things you’re going to consider when fixing it.

Developing a clear problem statement will help you get off to the right start. You want to try to come up with a clear statement of what you’re trying to solve, and then test this statement against a set of criteria

From Bulletproof Problem Solving:

“We test the problem definition against several criteria: that it is specific, not general, that we can clearly measure success, that the definition is bounded both in time frame and the values of the decision maker and that it involves definite action being taken. This step may appear constraining, but it leads to the clarity of purpose essential for good problem solving”

The well-known mnemonic SMART is handy for framing up your problem statements (keep them specific, measurable, action-oriented, relevant and timely). It’s worth spending some time properly defining the problem before you waste a lot of time “solving” it only to realise you’ve taken the wrong path. A problem definition worksheet like this blank one, or this example can help you frame up a problem and consider its main components.

Break it down

This is perhaps the step that sets this framework apart from others. Lots of problem-solving approaches talk about breaking problems down into component parts, but the Bulletproof approach recommends a specific, highly visual approach to doing this – they’re call logic trees.

A logic tree maps the components of a problem so you can easily visualise and disaggregate its different components or “drivers”. Issue trees are a great early step because they help you understand a problem properly and frame up the analyses that are going get you closer to a solution.

Different types of logic trees map problems differently. Mostly they are used to map the factors or components that make-up a problem (these can be displayed both inductively, or deductively). Logic trees can also be used to set out different hypotheses you want to test or to map different decisions you need to make in solving a problem.

If you’re interested in learning more about different types of trees and seeing examples, there are lots of blogs written to help people get jobs at places like McKinsey and they explain it well. Here is a detailed example.

The important thing to take from this is that visually mapping a problem puts you in a great position to tackle it systematically and holistically

From Bulletproof Problem Solving:

“Logic trees are really just structures made for seeing the elements of a problem in a clear way and keeping track of different levels of the problem which we liken to trunks, branches, twigs and leaves. You can arrange them from left to right, right to left, or top to bottom – whatever makes the elements easier for you to visualise… think of logic trees as a mental model of your problem.”

One thing that is important to get right in creating a logic tree is to ensure that it’s “mutually exclusive, but collective exhaustive.” That is you want to make sure that the “branches” of your tree don’t overlap, but at the same time, don’t miss any part of the problem.

From Bulletproof Problem Solving:

“Trees should have branches that are [both]:

Mutually Exclusive: The branches of the tree don’t overlap or contain partial elements of the same factor or component. This is a little hard to get your head around, but it means that the core concept of each trunk or branch of the problem is self-contained, not spread across several branches…


Collectively Exhaustive: Taken as a whole, your tree contains all of the elements of the problem, not just some of them. If you are missing parts, you may very well miss the solution to your problem”

Tom recently used this approach to break down the problem of student behaviour at our school. He began this process himself on a sheet of paper and then workshopped this “first cut” with his sub-school leadership team to refine and add bits that it missed. In all the process took around two hours, but it provided a really clear visual display of the problem, and the different things driving it. A tree-like this disaggregates a problem into more manageable “chunks” that can be appropriately divided amongst a team to test and develop solutions.

Click the picture to open it full-screen in another window

Prioritise the issues

Logic trees provide a visual display that make it easy to prioritise the critical things you need to work on.

From Bulletproof Problem Solving:

“Good problem solving is just as much about what you don’t do as what you do, and good prioritisation in problem solving makes solutions come fast and with less effort”

You do this by “pruning the tree.” Not all parts of your tree are equal, and it makes sense to chop some parts away so you can make the best of use of your resources. You don’t want to waste time trying to solve every “leaf” on your tree.

There is a neat tool for doing this. Using a 2×2 matrix (that’s consultant jargon for “small table”), you can consider the extent to which individual limbs, branches or leaves contribute to the problem and the extent to which they can be solved by the efforts of you and your team.

Below, you can see a few of the components of Tom’s classroom behaviour problem plotted against these variables. The team has considered the things that they can influence, given their role within the school. This means things that might be generally malleable, like curriculum, have been classified as “low ability to influence.” Sure, these things can be adjusted, but not by the team currently looking at the problem. You can also see that in this problem, the team will prune a lot of the “student factors” from the tree. This is not because these things don’t matter – they do – it’s just that relative to other parts of you don’t have as much influence on them right now.

Click the picture to open it full-screen in another window

This exercise helps you work out what you can do, and then what of those things you should do. The top right quadrant of this chart is where it’s at, really anything north-east of that dotted arc – these are the parts of the problem that have the largest potential, and they are things that you can affect.

Plan the work, do the work

So, you understand the problem, and now it’s time to fix it, right? Slow down speedy. It’s important to realise that “doing the work” here doesn’t mean solving the problem – that’s implementation, we’re still firmly in the “solution design” phase.

Remember that this framework is designed for complex problems that don’t have easy answers. With these problems, it’s worth spending time making sure that the solution is right before going ahead with implementing it. Too many change projects in schools skip or skim this step, and it can mean that you end up throwing a lot of mud at the wall, rather than executing well thought-out and appropriate solutions.

The key to this is good analysis. The logic tree and prioritisation activities you have engaged in up to this point have given you a good mental model of the problem and your hypotheses about what it’s going to take to solve it. Now you need to make sure this thinking is right – you do this by collecting information, analysing it and drawing conclusions to confirm or alter your hypotheses.

Before analysing everything in sight though, it’s worth planning your activities with a “work plan”. A good work plan helps you consider, what happens, why it needs to happen, who does it, and when it gets done.  Rather than investing a lot of time in a boring, detailed work plan, the Bulletproof approach suggests making it short, but highly specific to quickly focus your information gathering and analysis efforts

Again, you can see an example below of Tom’s behaviour problem:

Click the picture to open it full-screen in another window

The next step is to do the analysis. This involves collecting information and analysing it to draw conclusions about the validity of your hypotheses. In business and academia, there are lots of fancy tools that are used for this – Bayesian analyses, Monte Carlo simulations (neither as fun as the casino nor the biscuit), machine learning etc.

In most school problem solving, we don’t need this sort of analytical power. What we do need is a systematic approach to what data and information we’re collecting and what questions we’re expecting it to answer. The specific analyses you’ll need to conduct will vary from problem-to-problem. In some cases it might involve looking at empirical research, or collecting and analysing data from your school or community. In the behaviour example, the main analyses were around observing classroom practice to determine if processes, or practice were undermining behaviour.

Synthesise findings and tell a great story

Now, with analyses complete, you should have a coherent solution to starting to form. The first step in synthesising is to gather the outputs from your different analysis efforts and to pull them together into a coherent solution.  

Your logic tree can help with this, as you move towards a solution you can draw together the individual findings of the analysis work on each branch or twig of your tree to create an overall picture of how you will solve the problem at hand.

As a solution starts to form, the Bulletproof approach recommends a long-standing McKinsey communication technique – the Pyramid Principle. This approach suggests structuring communication like a pyramid, leading with a succinct statement of the highest level of the solution, and following with the supporting arguments that back up your top-level answer.

This approach avoids you taking stakeholders on a long “parade of knowledge” where they keep wondering what hidden agenda you’re trying to trap them in. Instead, you clearly and transparently provides the answer to the problem, then follow with the reasons why this is the best answer. A lot has been written on the Pyramid Principle that can help you use it to communicate your solution effectively. For more check out this book by McKinsey’s Barbara Minto or shorter things like this or this.

Key takeaways

  • Define the problem. Frame up a problem statement that is specific, measurable, time bound and considers the values of the decision maker.
  • Break it down. Use an “issue tree” to visually map the main components of the problem.
  • Prioritise the issues. Target your work on the high impact, highly impactable parts of the tree.
  • Plan the work. Do the work. Develop a highly specific plan of who does what.
  • Synthesise findings and tell a great story. Use the “pyramid” principle to sell your solution.